2015 December Extreme weather in the UK

Applying three independent methodologies of extreme event attribution, we show that temperatures and precipitation in the UK in December 2015 were extremely unlikely even in a warming world with observed SST patterns, including El Niño, as an additional driver. This indicates that random weather noise played a very large role in December’s weather. At the same time, the event was much less likely in the representations of a climate without human influence, showing that climate change greatly affected the odds of such a month occurring.

The observed temperature anomaly is so far outside the expected distribution that the odds are difficult to determine. We find that anthropogenic climate change approximately doubled the occurrence probability of the event for lower return times.  Analysis of the historical link between the observed CET dataset and El Niño shows no discernible influence on the CET in winter. This is confirmed by a coupled model analysis that only shows a weak connection. The weather@home simulations including all ocean temperatures are warmer than the Climatology ensemble. This includes El Niño, but also the warm subtropical Atlantic Ocean, which was the source region of the mild air flowing to Britain in December 2015.

Similarly all three methods show an increase in the likelihood of high precipitation in Northern English winters due to human-induced climate change. The connection with the El Niño signal is weak in December, but the weather@home simulations reveal an increase in the likelihood of very wet Decembers due to the ocean temperatures observed in December 2015.

What happened with the weather in December 2015?

The UK in December 2015 had an extreme month of weather. Central England had its warmest December on record, with an average of 9.7ºC being 5.0ºC above normal . Average rainfall in Northern England and South Scotland, at 255 mm/month was also anomalously high. The highest daily rainfall anomalies came with Storms, Desmond, Eva and Frank hitting the UK in December, causing extensive flooding and damage in Northern England.

Figure 1: Monthly average temperature and rainfall anomalies relative to the 1981-2010 average, from http://www.metoffice.gov.uk/climate/uk/summaries/2015/december. Contains public sector information licensed under the Open Government Licence v1.0.


Figure 1 shows temperature anomalies were highest in Southern England and lower, but still positive, further north. In contrast, rainfall anomalies in December 2015 show a clear north-south divide, with many parts of Northern England and Southern Scotland experiencing over three times normal monthly rainfall, while in Southern England rainfall was closer to normal. Hence in the rainfall series for England and Wales alone, December was not a particularly unusual month, while precipitation averaged over Northwest England, Northeast England and South Scotland showed the highest value for any month since the start of reliable observations in 1873.

Figure 2. GISTEMP global mean temperature and OSTIA observed SST anomalies for December 2015 relative to 1985-2013. This shows the strong El Niño pattern, as well as anomalies in the North Atlantic.


This period was also characterised by the highest global mean temperatures ever observed and a particularly strong El Niño. Meanwhile, sea surface temperatures in the Atlantic immediately to the west of the UK were unusually cold, and unusually warm further towards the US coast. This prompts the question: what role did anthropogenic climate change play in these extreme weather events relative to these natural fluctuations in sea-surface temperatures and/or the natural chaotic variability of the atmosphere?

We use a combination of statistical and dynamical modelling to assess the relative roles of observed sea surface temperature anomalies and anthropogenic influence on climate in these temperature and precipitation anomalies. The section Observational analysis gives the results from statistical modelling based on observed time series alone. The sections Large ensemble atmospheric simulations: Weather@home  and Coupled model simulations: EC-Earth give results from the weather@home model simulations and from the independent coupled model EC-Earth and explain the experimental setup of the two methods.